Re: opensmd fails to start on connectx-3 card
Hi, note that opensm can be run as a daemon -- are you sure there are no other instances running on the system? If no, then you may be missing some of the underlying userspace libraries needed for the...
View ArticleRe: opensmd fails to start on connectx-3 card
05:00.0 Ethernet controller: Mellanox Technologies MT27500 Family [ConnectX-3] Subsystem: Mellanox Technologies Device 0049 both ports are in ETH mode, however changing to ib mode fails with this...
View ArticleRe: opensmd fails to start on connectx-3 card
hmmm.... I just tried the connectx_port_config command on my HCA and it worked as shown below. what version of OFED are you running? You can type... [root@localhost ~]# ofed_info | head...
View ArticleRe: opensmd fails to start on connectx-3 card
MLNX_OFED_LINUX-2.0-2.0.5 (OFED-2.0-2.0.5): the card itself is connected to an ethernet switch, for RoCE, however this shouldnt preclude IB mode should it?
View ArticleRe: opensmd fails to start on connectx-3 card
My HCA is also connected to an Ethernet switch and I was able to run the connectx_port_config command. Can you provide the output of: # mstflint -d 05:00.0 q I want to check your HCA's PSID.
View ArticleRe: opensmd fails to start on connectx-3 card
Branko,Of course, here is the output. Image type: ConnectXFW Version: 2.30.3000Rom Info: type=PXE version=3.4.142 devid=4099 proto=ETHDevice ID: 4099Description: Node Port1...
View ArticleRe: opensmd fails to start on connectx-3 card
Well, I think that's unfortunately the issue. A PSID of MT_1080120023 means you have a ConnectX-3 Ethernet NIC (OPN of MCX312A-XCB) which I believe can only be configured for Ethernet and not...
View ArticleRe: opensmd fails to start on connectx-3 card
interesting. these cards are just EN, however IB support in the VPI line is required to achieve RoCE?
View ArticleClik here to view.
Re: opensmd fails to start on connectx-3 card
You should still be able to run RoCE on your Ethernet NIC. As long as you have OFED installed, which you do, you should be good to go. Here's a link to some documents that can assist you with RoCE...
View ArticleRe: Proper Configuration for IB-FDR and RoCE
Thanks, that's a good point.... I guess the OpenMPI stack things that ithas 2 phy transports and trying to load-share, runs into connectivityproblems. Do you think we can have both RoCE and IB active...
View ArticleRe: UFM 4.0 Monitoring History
Hi Yairi, My answer is late too, sorry for that.I'll try to use the csv export fonctionnality. The fact is that internal ufm display is really convinient for fast analysis. Maybe I'm missunderstanding...
View ArticleUFM experience on large scale cluster ?
Hello, As you may read on other thread, I'm running UFM on a small sized cluster ~200 nodes.Performance is ok for the gui and service, and also for monitoring history for a database located on remote...
View ArticleRe: UFM 4.0 Monitoring History
Putting together the numbers: server=2 ports (for each HCA) so 20 servers runs 40 ports (potentially). 36 ports switch is under the 40 limitation so no issues with that part. i guess, for your...
View ArticleRe: Can I use FDR and QDR on the same infiniband switch?
Wonderful. My resources telling me that 6012 is coming out really soon. should be perfectly OK to interop between QDR, FDR10, FDR. Good luck
View ArticleRe: Management command failed in KVM for SR-IOV
Gone.....? if i hear about anything around this area i will come back. let us know if it comes back or whether you can figure out a change you made that might have scare this issue away.
View ArticleRe: MHGH28-XTC firmware 2.9.1000 non-functional with ESXi 5.1?
Good catch - be aware that for hypervisors (VMWare, KVM, ZEN, etc.) for supporting SRI-OV you actually need a certain HW - newer servers would usually have the ASIC on-board to support SRI-OV. i am...
View ArticleRe: UFM 4.0 is GA !!
We have a licence for UFM 3.x do customers automatically get 4.0? From where can we download the software and where are the changelogs?
View ArticleClik here to view.
Re: Kernel Modules from Mellanox OFED Stack Won't Load
Bump Mellanox OFED 2.0-2.0.5 with kernel 2.6.32-358.11.1 (especially for Lustre 2.1.6) compat: exports duplicate symbol __pskb_copy (owned by kernel) Any update on this? You folks using Lustre 2.1.6...
View ArticleRe: Kernel Modules from Mellanox OFED Stack Won't Load
I'm in the same situation with 2.6.32-358.11.1.el6_lustre.x86_64 for Lustre 2.1.6. Any ETA on when a workaround will be ready?
View Article